The Nicotine Pregnancy Paradox: Why Some Doctors Are Prescribing Vapes to Expectant Mothers

Dr. Linda Bauld, a professor of public health at the University of Edinburgh, tells a story that illustrates the impossible position of the clinician treating a pregnant smoker. 'A woman comes in, 20 weeks pregnant, smoking a pack a day. She's tried patches—they don't work for her. She's tried cold turkey—she relapsed within a week. She's terrified, she's ashamed, and she's still smoking. What do you tell her? Continue smoking, which we know is harming her baby? Or try an e-cigarette, which we suspect is less harmful but can't prove? Either way, you're making a recommendation with incomplete evidence about a developing human life.' This is the nicotine pregnancy paradox—a clinical dilemma with no good options, only less-bad ones, that a small but growing number of clinicians are resolving in favor of vaping as harm reduction. The practice is controversial, ethically fraught, and possibly life-saving.

The evidence on which clinicians must base their recommendations is maddeningly incomplete—and the gaps are structural, not accidental. Pregnant women have been systematically excluded from clinical trials of e-cigarettes for the same ethical reasons they're excluded from most drug trials: the fear of harming a fetus makes researchers and ethics boards extremely conservative. The result is that we have randomized controlled trial data demonstrating that e-cigarettes help non-pregnant smokers quit, and observational data suggesting that e-cigarettes during pregnancy are less harmful than continued smoking, but no trial data directly comparing e-cigarette use to smoking or NRT in pregnant populations. The most-cited study, a 2022 trial in the UK that attempted to randomize pregnant smokers to vaping versus NRT, struggled with recruitment precisely because of the ethical concerns—midwives were reluctant to refer patients, and patients were reluctant to enroll. The evidence base the clinicians need doesn't exist, and the barriers to creating it are formidable.

The pharmacological reality complicates the clinical picture further. Pregnancy alters nicotine metabolism dramatically—the half-life of nicotine drops by roughly 50% during pregnancy as hepatic metabolism accelerates. This means that pregnant smokers may experience more intense and frequent withdrawal symptoms than when they weren't pregnant, making cessation harder precisely when the stakes are highest. It also means that standard-dose NRT, which was calibrated for non-pregnant adults, may under-dose pregnant women, reducing its effectiveness. Some clinicians have begun prescribing higher-dose or combination NRT during pregnancy, a practice that runs counter to standard guidelines but makes pharmacokinetic sense. The same logic applies, controversially, to pregnant women who switch to vaping: they may need higher nicotine concentrations than non-pregnant vapers to manage withdrawal effectively.

The ethical framework that some clinicians are using to justify recommending e-cigarettes to pregnant smokers is borrowed from other domains of medicine where ideal evidence is unavailable but a decision must be made. The principle is *harm reduction under uncertainty*: when the status quo (continued smoking) is known to be severely harmful, and the alternative (vaping) is plausibly less harmful based on the available toxicological and observational data, recommending the alternative can be ethically justified even in the absence of definitive evidence. This framework acknowledges the moral weight of uncertainty—recommending something that might cause harm is a grave responsibility—while also acknowledging that inaction is itself a recommendation. Telling a pregnant smoker 'we don't know enough about e-cigarettes to recommend them' without offering an effective alternative is, functionally, recommending continued smoking.

The fetal neurodevelopmental question is the hardest variable in the calculus. Nicotine crosses the placental barrier and has measurable effects on fetal brain development, including alterations to neurotransmitter systems, increased risk of attention and behavioral disorders in childhood, and potential long-term cognitive effects. This is true whether the nicotine comes from cigarettes, NRT, or e-cigarettes—nicotine is nicotine. The harm-reduction argument for vaping during pregnancy therefore rests on the claim that the non-nicotine components of cigarette smoke (carbon monoxide, which reduces fetal oxygen delivery, and thousands of toxicants with direct placental toxicity) are more damaging than nicotine alone, and that removing these while maintaining nicotine delivery is a net gain for fetal health. The claim is biologically plausible, supported by animal models, and unproven in human pregnancy. The uncertainty is deep and consequential.

The institutional response to this dilemma varies by country and reflects the broader schism in nicotine policy. In the UK, where vaping is integrated into the public health framework, the National Health Service's pregnancy smoking cessation guidance acknowledges that 'e-cigarettes are likely to be significantly less harmful than smoking' and does not prohibit their use during pregnancy, while stopping short of recommending them. In Australia, where vaping is medicalized and heavily restricted, the official guidance is unequivocal: pregnant women should not use e-cigarettes. In the United States, the guidance is cautious and leans toward recommending FDA-approved NRT over e-cigarettes, but the FDA has not taken a definitive position. Each of these institutional postures is defensible given the evidence. None of them solves the clinician's problem.

The resolution of the nicotine pregnancy paradox awaits better evidence—evidence that the research community has an ethical obligation to generate, despite the challenges. Pragmatic clinical trials that randomize pregnant smokers who have failed NRT to vaping versus continued support for abstinence are feasible, ethical (given that all participants are already exposing their fetuses to cigarette smoke), and urgently needed. Without such trials, clinicians will continue to make the most consequential recommendations of their practice—recommendations that affect two lives, not one—on the basis of extrapolation, plausibility, and hope. And pregnant smokers, the population with the most to gain from effective cessation, will continue to be denied the evidence-based options that every other smoking population increasingly takes for granted. As Dr. Bauld puts it: 'We keep saying we need more evidence before we can recommend anything to pregnant smokers. But they're not waiting. They're smoking, right now. And every day we wait is a day we've chosen, by default, to recommend the thing we know is killing them.'